Changing the energy of an immune response

Abstract

The breakdown of nutrients into the critical energy source ATP is the general purpose of cellular metabolism and is essential for sustaining life. Similarly, the immune system is composed of different cell subsets that are indispensable for defending the host against pathogens and disease. The interplay between metabolic pathways and immune cells leads to a plethora of different signaling pathways as well as cellular activities. The activation of T cells via glycolysis-mediated upregulation of surface markers, for example, is necessary for an appropriate effector response against an infection. However, tight regulation of immune cell metabolism is required for protecting the host and resuming homeostasis. An imbalance of immunological metabolic function and/or metabolic byproducts (reactive oxygen species) can oftentimes lead to diseases. In the case of cancer, overactive glucose metabolism can lead to hyperproliferation of cells and subsequent decreases in cytotoxic T cell activity, which attack and destroy the tumor. For this reason and many more, targeting metabolism in immune cells may be a novel therapeutic strategy for treatment of disease. The metabolic pathways of immune cells and the possibilities of immunometabolic therapies will be discussed.

Aerobic respiration

Cellular metabolism is necessary for generating energy and sustaining life. Through a series of steps involved in glycolysis (glucose), fatty acid oxidation, and amino acid (protein) oxidation, cells can break down ingested products into critical energy sources. This energy, better known as adenosine triphosphate or ATP, is synthesized as a result of the degradation of nutrients. Oxygen (O₂) plays a key role in enabling reactions required for the formation of ATP. In human cells, oxidative phosphorylation is the main process leading to the generation of ATP [1]. The degradation of nutrients through the glucose oxidation, fatty acid oxidation, or amino acid oxidation pathways converge to all produce acetyl-CoA, a key molecule that provides a carbon source for fueling the tricarboxylic acid (TCA) cycle. The oxidation of acetyl-CoA to carbon dioxide (aerobic respiration) then allows for the subsequent reduction of nicotinamide adenine dinucleotide (NAD+) and flavin adenine dinucleotide (FAD) via the TCA cycle. The intermediate products, NADH and FADH₂, serve as electron transport chain coenzymes for oxidative phosphorylation. For efficient respiration, electrons must be transferred from NADH and FADH₂ to oxygen via the mitochondrial complexes along the electron transport chain within the inner mitochondrial membrane. Electrons must be strictly allocated down the electron transport chain while protons must be pumped across the mitochondrial membrane. This movement of electrons not only facilitates the production of H₂O, but also drives a proton gradient that causes the phosphorylation of adenosine diphosphate to adenosine triphosphate. ATP can then be utilized for a number of events including DNA/RNA/protein synthesis, cell signaling, cytoskeletal rearrangement, cell proliferation, and metabolic pathways. Despite the necessity for energy production, the electron transport chain is also responsible for the formation of mitochondrial reactive oxygen species (ROS) through continuous ‘leakage’ of electrons, causing partial reduction of O₂ molecules [2]. Such events lead to the generation of superoxide (O₂ ^-), hydrogen peroxide (H₂O₂), and hydroxyl radical (OH^-), which can both stimulate physiological actions as well as damage important molecules in the body [3], depending on the quantity. At a low abundance, ROS help control normal cellular functions, as demonstrated by the immune system’s reliance on this type of signaling for regulation, activation, T cell proliferation [4,5], NF-κB activation [6], and signal transduction [7-9]. Conversely, high levels of ROS lead to oxidative stress, which has been linked to a variety of diseases, aging, and cell death [10-13]. ROS production is, thus, a necessary evil for functional aerobic metabolism.

Aerobic glycolysis

Although aerobic respiration is the main source of ATP in most mammalian cells, an alternative form of metabolism, glycolysis, is crucial in both immunity and disease states. Glycolysis enables the conversion of one glucose molecule to 2 pyruvate molecules, with subsequent production of lactate, NAD+, and ATP. The utilization of glycolysis is a less efficient form of metabolism. During mitochondrial respiration, a cell is able to produce 38 ATP molecules; however, during glycolysis, only 2 molecules of ATP are generated. The obvious 19-fold increase in ATP via oxidative phosphorylation seems more advantageous to the cell, especially since both metabolic pathways can use glucose as the starting material. However, glycolysis is especially important in times of hypoxia (anaerobic) and can occur even in the presence of oxygen (aerobic), such as with tumors and immune cells. Both anaerobic and aerobic glycolysis are crucial for the maintenance of tumor cells [14-16]. Tumor cells switch their energy production from oxidative phosphorylation to glycolysis upon transformation to malignancy [17,18]. This enables the tumor to rapidly grow in hypoxic environments and evade host immune cell defense mechanisms [19-21]. In oxygenated environments, tumor cells display augmented glucose transport and glycolysis [22,23]. This counterintuitive metabolic programming has been attributed to over-adaptation to hypoxic environments, a greater need for macromolecules during unrestrained proliferation (glycolysis drives better protein/nucleotide synthesis), and elevated expression of the glycolytic enzyme hexokinase [24,25]. This high aerobic glycolysis seen in cancer cells is called the Warburg effect [14,26] and has led to the development of anti-glycolytic drugs for cancer treatment [27,29]. During proliferation, T cells and cancer cells show similar metabolic programming. Thus T cells, even in the presence of sufficient oxygen, also choose to ferment glucose, as further discussed below.

Immune cell metabolism

The immune system is made up of two different arms: the innate and adaptive immune system. The innate system is a first-line of defense against pathogens and foreign substances. Unlike the adaptive arm, the innate response is non-specific and is mediated by antigen-presenting cells (APCs) as well as granulocytes. Examples of innate cells include dendritic cells, macrophages, and neutrophils. After successful priming by the innate cells, the adaptive immune response, made up primarily of T and B cells, provides antigen-specific protection against the insult, either through the release of cytotoxic granules, cytokines, or antibodies. Immune cells, like most other cells in the body, utilize nutrients via cellular metabolism. At rest, immune cell metabolism is able to regulate cell volume, ion integrity and growth [30]. However, in addition to housekeeping proliferation and sustenance, ATP within the immune cells must be ready to carry out various functional activities such as phagocytosis, activation, antigen presentation/processing, migration, phosphorylation, differentiation, and effector responses [30,31]. Most of these actions are thermodynamically taxing, requiring notable and rapid changes in metabolism [30,32]. Furthermore, immune cells must facilitate cytoskeletal changes, increased ion signaling, enhanced phospholipid turnover, and greater macromolecule synthesis in a very short time during rapid energy consumption [33]. Importantly, resting immune cells, especially those of the adaptive arm, contain little glycogen stores, resulting in the dependence of imported glucose to uphold metabolic needs [34-36].

Because of the diverse functionality of immune cells, several important differences exist between their metabolism and that of other cells within the body. Alveolar cells, for example, are reliant primarily on oxidative phosphorylation for the generation of sufficient ATP [37]. Those cells which are in constant contact with oxygenated blood are especially formulated for mitochondrial respiration. On the other hand, immune cells travel through the body to monitor the peripheral tissues. Once a foreign antigen is detected, APCs migrate to the draining secondary lymphoid organs where they can process and present the antigen to lymphocytes. APCs are present in strategic areas of the body as resident phagocytes. These cells, therefore, are not necessarily exposed to normoxic conditions at all times. For instance, epidermal dendritic cells reside within the deep tissue layers of the skin [38], where oxygen tension is lower than the dermis [39]. Such conditions create a slightly hypoxic environment in which key immune cells must be able to utilize alternate forms of metabolism in order to survive and function properly [40]. Similarly, lymphocytes, upon activation within the secondary lymphoid organs, travel to the site of inflammation by traversing the endothelial cell wall into the target area [41]. Once again, the migration of lymphocytes away from the source of oxygen causes slight hypoxia and a resultant loss of dependence on oxidative phosphorylation. Most sites of inflammation are also areas of lowered oxygen, with innate phagocytes clogging the blood vessels [42-44]. Localization of immune cells thus requires adaptation to different oxygen levels and promotes more glycolytic pathways [45,46].

Although influential, the environment is not the only element dictating the metabolic choices within immune cells. The activation of both innate and adaptive immune cells is absolutely critical for protecting the body from pathogens and insults. Consequently, the cells cannot afford to be inefficient in their nutrient metabolism. With this being said, it would seem likely that immune cells should generate ATP via oxidative phosphorylation, fostering the most energy from the nutrients provided. However, this is not entirely the case. Although some mitochondrial respiration does occur in immune cells, the level at which it is used depends significantly on the cell’s specificity and state of reactivity. Activated (and some inactive) immune cells prefer to utilize glycolysis, as it is 100-times faster than oxidative phosphorylation for macromolecule synthesis and proliferation [47].

As mentioned above, naïve APCs monitor the body for foreign substances. In this resting state, myeloid cells and granulocytes favor glycolysis [48,49]. Homeostatic protein turnover, degradation, and synthesis all occur via growth factor utilization [50,51]. Once antigen is phagocytosed, APCs immediately upregulate costimulatory molecules and process and present antigen on their cell surface. Such events require greater ATP; however, APCs retain their dependence upon glycolysis [48,49]. Dendritic cells, for example, are known to undergo metabolic changes towards greater glycolysis upon toll-like receptor (TLR) stimulation [52]. Moreover, classically activated macrophages (known as M1), which promote proinflammatory cytokines, are known to be regulated by glycolysis; however, alternatively activated macrophages (M2), which are anti-inflammatory, rely more so on mitochondrial respiration [53,54]. Glycolysis is, therefore, important for inflammatory responses and can be detrimental in the face of chronic inflammatory diseases. Granulocytes, such as neutrophils, also favor glycolysis [48,55]. Neutrophils are the first mediators at the site of foreign entry. Their immediate degranulation and pyrogenic secretion lead to subsequent cell death [56]. In accordance with the glycolytic reliability of APCs, quick responses are needed by neutrophils, but survival via greater ATP production is not necessary. Fast reactivity and turnover rates of both APCs and granulocytes, therefore, make sense with their choice to generate ATP through the more rapid glycolytic pathway.

Lymphocytes, on the other hand, rely heavily on oxidative phosphorylation during resting states but switch their metabolic needs to glycolysis during activation (Figure 1). Subsequently, some lymphocytes return to oxidative phosphorylation after clearance of an antigen to generate memory [57,59]. Therefore, there is a cyclical pattern of metabolic pathways, fluctuating between anabolism and catabolism, as reviewed by Pearce et al. [60]. At rest, T cells need a constant supply of nutrients as well as TCR stimulation [61]. The idea of TCR ‘tickling’ has long been thought to prevent deletion of mature T cells [62]. The active metabolism behind the quiescent state is a relatively novel idea, with several transcription factors implicated in regulating this process [63]. Circulating naïve lymphocytes undergo oxidative phosphorylation to generate a surplus of ATP reserves via catabolic metabolism, the breakdown of nutrients [36]. In a sense, quiescent immune cells remain ‘at attention’ in order to quickly mobilize following antigen stimulation [64]. Preservation of quiescence is mediated by turnover of cell cycle proteins, a very active event which requires a lot of ATP [50,51], and by upregulation of cyclin-dependent kinase inhibitors [65]. Quiescent cells not only utilize glucose, amino acids, and lipids for ATP generation, but can also extract nutrients from those proteins which are degraded, via autophagy or self-eating [30,66,67]. Engagement of the TCR as well as growth factors and homeostatic cytokines, like IL-4, IL-5, IL-7, and IL-3 [68-70], also all play roles in keeping naïve cells alive. Cytokine-receptor signaling, specifically, can activate protein kinases that are necessary for the uptake of adequate ATP to preserve homeostatic processes [36], whereas a lack of TCR interaction will downregulate glucose transport, ATP, and mitochondrial potential [61]. Insufficiencies in glucose uptake by T cells will lead to BAX induction and apoptosis [71]; however, this stringency serves to control the naïve T cell population, ensuring the turnover of existing cells as new cells are produced to avoid over accretion [61].

Figure 1.

Lymphocyte metabolism fluctuation during resting and activated states. Quiescent, or resting, lymphocytes primarily utilize oxidative phosphorylation to build up reserves of ATP in preparation for activation. Activated, or proliferating, lymphocytes predominately use aerobic glycolysis due to its rapid speed and critical role in forming biosynthetic precursors.

Like the innate cells, lymphocytes need to become rapidly activated upon antigen stimulation. Such changes in metabolism are well-documented in the case of mitogen-stimulated lymphocytes [60,4,72]. Within the first 24h post-mitogen stimulation, lymphocytes considerably enlarge their size [60]. During this time, new macromolecules are being synthesized, including nucleotides and proteins. Following the growth, T cells then divide every 4-6 hours [72] and eventually will differentiate into effector cells. Effector functions such as cytokine production and cytotoxic granule release then enable the cell to attack the target area. This quick change in cell size and function relies primarily on obtaining nutrients from the environment and driving glycolysis [36]. Instead of initiating the long process of oxidative phosphorylation, T cells convert glucose to pyruvate via the more rapid glycolysis [47], and in the process, generate ATP and lactate, which allows for the conversion of NADH back to NAD+ to retain glycolysis. The dependence of T cells on glucose is so great, that even in the presence of excess glutamine, which is another carbon source that can be metabolized by lymphocytes, proliferation is stunted [68]. T cells are not able to enhance mitochondrial respiration to a level that meets their energy needs. Moreover, T cells require high levels of NADH as macromolecular precursors; therefore, greater lactate production equals more NAD generation [32]. Notably, following mitogen-stimulation, an excess of lactate can be measured from T cells [73,74]. Aerobic glycolysis ensures that enough energy is made to propel macromolecule synthesis (anabolic metabolism), which is ultimately crucial for clearance of a pathogen, as well as to keep the cells alive and functional [36,75,76]. Some oxygen consumption does still occur [77], yet is typically only utilized in situations where glucose is limited [78,79]. Mitogenic-stimulation of peripheral blood mononuclear cells, for example, causes oxygen consumption attributed to ATPase activity, protein synthesis, and nucleic acid synthesis [31,64]. Moreover, activation of lymphocytes promotes a calcium flux, which will also drive the upregulation of mitochondrial enzyme activity for sufficient oxidative phosphorylation [80,81]. In the case of infection, T effector cells promote clearance of the pathogen and primarily utilize glycolysis for rapid growth. By the time pathogen has been cleared, the mitochondrial capacity of effector cells is reduced in such a way that they can only sustain viability through glycolysis. In the absence of adequate nutrients and IL-2 present during infection, effector cells are unstable and eventually will undergo apoptosis. Certain clones, nonetheless, will retain their ability to switch back to oxidative phosphorylation. Such cells will contain greater mitochondrial mass, either through differences in biogenesis or asymmetric division [57]. In autoreactivity [82] as well as graft-versus-host disease [83], chronically stimulated T cells rely on oxidative phosphorylation in contrast to acutely activated cells (those discussed above) which depend on glycolysis. Oftentimes, such illnesses and metabolic outcomes correlate with mitochondrial dysfunction and/or increased mitochondrial mass present during the disease [84]. Lipid oxidation, specifically, is important for the generation of both regulatory T cells (Tregs) [85] and memory CD8+ T cells [86], countering glycolytic effector CD4+ and CD8+ T cells. Following T cell activation, not only will the clearance of antigen help generate memory, but remaining T effector cells will die via apoptosis due to decreased growth factors and metabolism [15]. Memory T cell metabolism is similar to naïve, in that both populations require oxidative phosphorylation; however, some critical changes occur over the course of an immune response to ensure memory T cell survival. Those that become memory cells will have higher TCR affinities, which allows for advantageous survival in conditions of nutrient limitation and low homeostatic engagement [61]. Furthermore, CD8 T cells, described to have substantial spare respiratory capacity after clearance of infection, will be long-lived memory cells [59]. Spare respiratory capacity is especially critical for producing sufficient ATP under times of stress and for boosting long-term cell survival [87,88], which would be necessary for memory response. Similarly, CD4 T cell memory is generated via blockade of a gene called Noxa that is responsible for driving apoptosis under conditions of limiting glucose [89,90].

Connecting immune signaling and metabolism

Stimulation of T cells via the TCR requires proper engagement by MHC-peptide on the APC. Additionally, costimulation during T cell activation is critical for downstream signaling and effector function. Conversely, a lack of costimulation can lead to T cell anergy and deletion [91,92]. Appropriate T cell activation is not only governed by mere mechanistic interaction and a cascade of signaling molecules, but there are important links to metabolism. CD28 is the quintessential costimulatory molecule for T cell activation. Its ability to bind CD80/CD86 on APCs enables downstream signaling and promotes T cell differentiation [93,94]. Analogous to insulin-receptor signaling, CD28 has been shown to enhance glucose metabolism by triggering an accumulation of glycolytic intermediates [35], stimulating glycolysis, and increasing glucose transporter expression [95]. CTLA-4, on the other hand, offsets the effects of CD28, reducing glycolysis and rendering cells quiescent [96]. One of the most essential downstream signaling cascades bridging the gap between T cell activation and metabolism is the PI3K-Akt-mTOR pathway. CD28, IL-2 and TCR engagement leads to PI3K-dependent Akt activation, which in turn increases the amount of glucose transporters on the plasma membrane as well as elevates activities of glycolytic enzymes i.e. hexokinase and phosphofructokinase [32,36,70]. PI3K as well as MAPK and NF-κB can all activate Myc, which is responsible for inducing glucose transporters as well as glycolytic enzymes [97,98]. Akt, in conjunction with STAT5, also plays a role in glucose uptake in resting T cells [99]. Unlike other cell types, lymphocytes only express the Glut1 glucose transporter [32,61]. In the absence of adequate TCR and/or cytokine stimulation during both the resting and activated states, Glut1 will be internalized, leading to downregulation of surface expression, reduced transport of glucose across the plasma membrane, and decreased viability of the cell [100]. CD28-mediated Akt signaling is especially important for glucose uptake as it is necessary for expression and trafficking of Glut1 to the cell surface [35,101]. Comparable to T cells, B cells also increase Glut1 expression following BCR engagement [34]. PI3K similarly plays a critical role in B cell proliferation and immunoglobulin synthesis through regulation of glycolysis and Glut1 [102]. Although Glut1 expression is critical for lymphocyte activation, a balance must exist; otherwise, overexpression of Glut1 can manifest into hyperactive lymphocytes and pathologies [101].

mTOR is another critical regulator of metabolism in immune cells [103]. The ability of mTOR to sense nutrient availability [104,105] leads to the induction of mRNA translation and protein synthesis [106]. Without proper mTOR signaling, T cell proliferation will be blocked [107] and anabolic storage processes will be decreased [108,109]. Inhibition of the PI3K-Akt-mTOR pathway can also lead to T cell anergy [110,111], whereas mTOR-deficient T cells do not differentiate into effector T cells, but instead regulatory T cells [112]. Additionally, mTOR has been linked to chemokine-dependent signaling, resulting in T cell migration [113-115] and cancer metastasis [116,117]. These data highlight the importance of proper metabolic signaling in initiating an effective adaptive immune response as well as reveal targets for therapeutic intervention.

Cytokine binding and cytokine receptor expression have also been connected with metabolism. Although immune cells primarily utilize glycolysis, some oxidative phosphorylation still occurs and is necessary for functionality. If mitochondrial respiration is blocked at different complexes of the electron transport chain, both TNF binding to its receptor on cells [118] and IL-2R expression on lymphocytes are reduced [5]. Interestingly, TNFα-deficient mice are actually protected from obesity-induced insulin resistance, highlighting the importance of proinflammatory cytokines in metabolic signaling [119,120]. Other innate immune cytokines, such as IL-1, IL-6, IL-3 and IL-7, also contribute to metabolism. IL-1 can prevent fatty acid synthesis [121], whereas IL-6 can both increase the levels of lipid and glucose metabolism [122]. IL-3, which is known to support the growth of myeloid and lymphoid cells, is important for sustaining Glut1 on the surface of lymphocytes [70] and has been directly shown to shift metabolism from oxidative phosphorylation to glycolysis [123]. In order for activated Akt to sustain glucose uptake in both resting and activated T cells, IL-7 must be present [68]. Overall, each cytokine binds a specific receptor and coordinates T cell function with metabolic needs.

Similar to macrophages, adipocytes can release adipokines (i.e. IL-1, IL-6, IFNγ, TNFα, MCP1) to also bridge the gap between immunity and metabolism [95,124,125]. Adipokines can recruit monocytes and lymphocytes into the adipose and promote proinflammatory and anti-inflammatory functions. Overnutrition can induce adipocyte hypertrophy, creating a hypoxic core and MCP-1 production, which facilitates macrophage entry into the adipose [126]. Lymphocytes associated with adipose tissue are oftentimes modulators of the infiltrating macrophages [127]. For example, Tregs are present in greater abundance in the adipose of lean mice, correlating with an anti-inflammatory macrophage phenotype [128]. Furthermore, proinflammatory effector T cells have been detected in the fat of obese mice, leading to the recruitment of even more proinflammatory macrophages [129] and contributing to insulin resistance [130,131].

Hormone secretion, from the adipose as well as other tissues, is also important in regulating lymphocyte function. Leptin, a hormone released from the adipose, along with insulin, which is secreted from the pancreatic beta cells, both play critical roles in connecting metabolism to the immune system. Leptin, which regulates food intake by inhibiting appetite, is low in times of starvation, resulting in decreased metabolism to maintain vital organs. Consequently, low leptin levels lead to immunosuppression [95]. In a well-nourished environment, leptin can modulate both the innate and adaptive arms of the immune system to promote greater cytokine production [132], decreased apoptosis [133] and skewing of T cells towards the T_H1 lineage [134,135]. The effects of leptin on T_H1 responses have been especially documented in the context of autoimmunity. Leptin has been shown to accelerate type 1 diabetes onset in NOD mice via enhancement of IFNγ-producing T cells [136]. Furthermore, higher leptin levels have been detected in female animals that are susceptible to EAE induction versus resistant males, positively correlating with an increase in T_H1 responses [137]. Heightened immune responses following leptin signaling may be damaging in autoimmunity; however, in obesity, failure of proper immunity resulting in increased infections has been linked to greater leptin insensitivity in severely overweight individuals [138]. In general, the immune system cannot function properly in times of over or under-nutrition [139], relating the many intricacies between metabolism and immunity. Similarly, insulin, which promotes cellular metabolism by stimulating the uptake of glucose and storage as glycogen, can play a role in modulating the T cell response. Beyond acting as a lymphocyte-specific antigen in type 1 diabetes, insulin helps shape T cell growth and function. Upon activation, insulin receptor is expressed on T cells. Insulin signaling then facilitates glucose uptake, amino acid transport, lipid metabolism, and protein synthesis [140]. Stimulating CD4+ and CD8+ T cells in the presence of insulin can induce more T_H2-type cells and cytokines [141], leading to a more anti-inflammatory environment. This data strongly suggests why lack of insulin signaling in both type 1 and type 2 diabetes can lead to both enhanced T_H1 cells and uncontrolled inflammation [142].

Modulators of metabolism and potential immunometabolic therapeutic implications

Mitochondrial activity has been implicated as a cause of aging, and metabolic dysfunction and ROS production have been linked to neurodegeneration, cancer, and autoimmunity [10-13]. An accumulation of ROS and redox-damaged byproducts eventually leads to cell dysfunction and death [143]. Indeed, mutations of the electron transport chain can diminish ROS production and thus elongate life [144,145]. Although immune cells contain higher levels of antioxidants than other cells [146] and rely on both glycolysis and respiration, aging immune cells show accrued impairment, causing reduced lymphocyte proliferation. Functional decline of immune cells, or senescence, oftentimes correlates with age, as free radical production overwhelms antioxidant defenses and the risk of infections/tumors is enhanced [147-149]. During a normal mammalian lifetime, metabolic pathways are kept in check via a number of endogenous mediators such as hypoxia-inducible factor-1 and uncoupling proteins, which prevent oxidative phosphorylation by partially dissipating the mitochondrial proton gradient. Pertaining to aerobic respiration, the existence of antioxidants particularly protects against oxidative stress and damage. Additionally, endogenous mechanisms do exist to restrict ROS production so as not to damage neighboring tissues. For example, Kupffer cells, macrophages residing within the liver, do not undergo respiratory bursts, thus protecting the surrounding parenchyma from any ROS-mediated destruction [150,151]. Conversely, peritoneal macrophages, which are more involved in clearance of infection, can experience an oxidative burst, with less threat of damage to surrounding tissue [152]. Despite these many mechanisms, improper metabolism of immune cells can oftentimes result in disease. The strict dependence of immune cells on glucose for survival and activation, however, may make them good targets for metabolic therapeutics [60]. Such therapeutics could potentially better control autoimmunity, transplantation rejection, neurodegeneration and cancer.

Hypoxia-inducible factor-1

Hypoxia-inducible factor-1 (HIF-1) is especially important for modulating metabolism in times of low oxygen conditions. HIF-1 is able to inhibit the progression of pyruvate into the TCA cycle by redirecting it to lactate production via activation of pyruvate dehydrogenase kinase [153-155]. Furthermore, HIF-1 can induce glycolytic enzymes while reducing mitochondrial oxygen consumption [155]. Such a switch exists to preserve the viability of cells in times of hypoxia. Interestingly, a similar change occurs in activated lymphocytes, as discussed above. The metabolic similarity between hypoxic cells and lymphocytes begs the question of whether or not HIF-1 plays a role in modulating T cell activation. Hypoxic areas within the body create a need for immune cells to survive and function properly in all environments, hence the importance of HIF-1 activation [40]. Under hypoxic conditions and anaerobic glycolysis, specifically during wound healing, T cells will shift from T_H1 to T_H2 type responses, directing a more anti-inflammatory function in the absence of oxygen [156]. In mice prone to type 2 diabetes, decreased levels of HIF-1 indeed lead to impaired wound healing [157].

Besides hypoxic conditions, HIF-1 is able to induce expression of genes that improve immune cell viability during aerobic glycolysis [75]. The expression of HIF-1 is initially provoked by insulin, IGF1 (insulin-like growth factor 1), and angiotensin, all of which play roles in growth and survival [158,159]. HIF-1 is increased in activated T cells and promotes expression of Glut1, aiding in T cell survival [40,160]. In addition to its importance in maintaining T cell viability, HIF-1 also helps regulate T cell subset differentiation. T_H17 cell differentiation requires enhanced glycolysis and expression of the transcription factor RORγT, both of which are increased via HIF-1 activation [161,162]. Furthermore, HIF-1 is known to directly repress Foxp3, the transcription factor critical for Treg induction [161]. Tregs, unlike other T cell subsets, are primarily powered through lipid oxidation [85]. Likewise, lipid metabolism, which would mainly utilize oxidative phosphorylation, can inhibit glycolytic-dependent T_H17 differentiation [163].

HIF-1 also plays important roles in controlling innate cell functions [153]. ATP, glycolytic enzymes and Glut1 expression are all regulated by HIF-1 in macrophages and neutrophils. Under hypoxic conditions, APC phagocytosis and antigen presentation as well as granulocyte responses are weakened [164,165]. Upon HIF-1-deficiency, innate cell motility, invasiveness, pathogen killing and T cell-stimulating abilities decrease even more so [49,166]. Without effective HIF-1 expression, APCs and granulocytes suffer dysfunctional host defenses. On the contrary, chronic inflammation and HIF-1 may together instigate tissue fibrosis, autoimmunity and tumor progression by affecting both innate and adaptive immune cells.

In chronic kidney disease and obesity, HIF-1 can switch from its proangiogenic function to promote fibrosis [167,168]. In relation to HIF-1’s ability to modulate T_H17 differentiation, mice deficient for HIF-1 are resistant to the inducible experimental autoimmune encephalomyelitis (EAE) [161,162], a rodent model of multiple sclerosis in which disease is largely mediated by T_H17 cells. Furthermore, HIF-1 has been shown to play major roles in prostate cancer tumorigenesis [169], breast cancer prognosis [170], and many other cancer outcomes [171], through induction of genes responsible for cell proliferation, angiogenesis, survival, migration, and glucose metabolism [172]. Immune cells play crucial roles in mediating appropriate wound healing, tolerating self-antigens, and cytotoxic killing of tumor cells. Therefore, blocking HIF-1 may allow for appropriate immunity and alleviation of disease. The list of HIF-1 inhibitors is expanding [171]. For example, digoxin, which inhibits HIF-1 gene and protein expression, can block tumor growth [173] as well as RORγt-dependent T_H17 differentiation [174], yet does not affect other T cell lineages. It is tempting then to speculate that while blockade of HIF-1 may shunt proliferation and T_H17 cells, cytotoxic CD8 T cells may still be active to allow for killing of tumors. Conversely, since HIF-1 is known to repress Treg differentiation, inhibition of this molecule may augment suppressive T cells. In the context of autoimmunity, this side effect may be beneficial in protecting against self-antigen recognition. However, in the case of cancer, combinatorial therapies of digoxin along with chemotherapy and radiation may be necessary for complete regression.

Uncoupling proteins

For oxidative phosphorylation to occur properly, collaboration between electron transfer and proton pumping is a necessity. A disturbance in the ‘coupling’ of electrons to protons would lead to an increase of futile proton current, decreased ATP production, and diminished ROS levels. In the context of immune cell mitochondrial dysfunction, such as in Alzheimer’s and diabetes, an interruption of oxidative phosphorylation may be beneficial for reduction of ROS byproducts. Endogenously, certain proteins exist to manifest this disruption. Uncoupling proteins (UCP) are known proton uniporters that, in the context of a proper activator, can uncouple mitochondrial respiration in a controlled way [175,176]. Such processes are used for thermogenesis from brown adipose tissue [177] and for reducing the production of free radicals from mitochondria [178]. UCP2 has been suggested to decrease pyruvate entry into the TCA cycle [179,180], overall limiting ROS production and age-related damage [181-184]. Overexpression of UCP3 in a high-fat diet fed mouse was able to rescue insulin signaling [119] and knockout of UCP2 drives persistent NF-κB activation as well as heightened ROS production in immune cells, resulting in resistance to certain infections [185,186]. These data reveal UCP2 as a plausible immunometabolic therapeutic target. T cells produce high levels of mitochondrial UCP2 following activation; this has been attributed to the need for rapid proliferation via glycolysis as well as the necessity of low level ROS production for adequate gene expression and signaling activity [187]. In an oxidative stress environment, where interplay between innate immune cells and T cells generates high ROS, uncoupling the electron transport chain may be useful in resetting homeostasis. Moreover, UCP2 decreases glucose-stimulated insulin release [188], highlighting its potential for preventing the release of self-antigen and thus controlling autoreactive T cell responses. To date, uncouplers have not been utilized in the treatment of autoreactive T cells. However, chemical uncouplers do demonstrate abilities in reducing oxidative stress. Rottlerin, a mitochondrial uncoupler, can reduce apoptosis of alveolar macrophages in a model of systemic autoimmune disease [189]. A study utilizing a mitochondrial fission inhibitor, which led to greater uncoupling, normalized oxidative stress levels in hyperglycemia [190]. Another drug, 2,4-dinitrophenol, leads to uncoupling through dissipation of the proton gradient, resulting in decreased hepatic insulin resistance in a non-alcoholic fatty liver disease model [191]. Furthermore, 2,4-dinitrophenol has been shown to enhance the adhesion phenotype (increased collagen and VEGF) for post-peritoneal surgical wound healing [192], a process in which macrophages play an important role [193]. Uncouplers targeting specific cells, therefore, may be a potential therapeutic for immune diseases where oxidative stress is high, whereas uncoupling inhibitors may be utilized for cancer, especially those which are resistant to chemotherapy [194,195].

Nutrient limitation

The immune system is highly dependent on the glucose levels available. Everyday physiological nutrient limitation throughout the periphery protects from the over accumulation of naïve T cells, allowing for the turnover of older cells and development of new [61]. Likewise, the adaptive immune response also relies on the availability of sufficient amino acids. Innate immune cells can control the supply of amino acids and thus, regulate T cell responsiveness. Upon CD40 ligation or LPS stimulation, APCs can increase their cysteine production and share with interacting T cells, which cannot make their own [196,197]. Cysteine is critical for T cell survival due to its necessity in glutathione production [198,199]. Similarly, indoleamine 2,3-dioxygenase (IDO) expression by macrophages can reduce extracellular tryptophan levels, which are also needed by, but not directly produced by T cells. A lack of tryptophan will eventually cause T cell anergy or apoptosis [200,201], again suppressing an immune response.

In a similar manner, environmental nutrient limitation may also affect an immune response. Caloric restriction causes greater lipid metabolism, lessening the dependence on glucose and thus decreasing an immune response [202]. Lipid metabolism is the chief mechanism of energy production in immune-privileged sites, such as the eye, brain, and placenta. Fatty acid utilization leads to lower costimulatory molecules, resistance to apoptosis, and less damage by free radicals [202-204], correlating with blunted immune responses. In contrast, immune-sensitive areas that depend on glucose metabolism are more susceptible to infection and death following an effective immune response [69,71,205]. Notably, saturated and unsaturated fatty acids also differ in their abilities to stimulate an immune response. Saturated fatty acids, which are considered more detrimental to health, induce greater activation of TLR2 and TLR4 on myeloid cells, whereas unsaturated fatty acids can inhibit TLR signaling and NF-κB activation [206-208]. Particularly, polyunsaturated acids can alter the T cell membrane, negatively impacting signaling and activation of lymphocytes [209,210]. Moreover, less caloric intake also correlates with better DNA repair, reduced antioxidant decline, diminished cancer rates, and an increase in lifespan [211-214]. In autoimmunity and metabolic syndrome, such confines may be highly effective in quelling inflammation [215,216]. Specifically, obesity has been associated with dysfunctional phagocytosis and respiratory burst in macrophages [217,218]. Similar to leptin insensitivity, in obesity, continuous ingestion of saturated fatty acids eventually will lead to a reduction in innate and adaptive immune responses, making individuals more susceptible to cancer and infections [219] and reducing wound healing capacity [220,221]. Caloric restriction of obese mice, which were at-risk for breast cancer, significantly reduced tumor growth and mimicked mTOR inhibition for regulating cell proliferation [222]. Nutrient limitation via endogenous and exogenous factors may therefore benefit a number of immune diseases with metabolic components.

Anti-glycolitics

Because of the high dependence of immune cells on glucose metabolism, anti-glycolytics have been implemented to limit immunity and treat disease. One of the most studied drugs is rapamycin. Rapamycin is able to inhibit glucose metabolism via blockade of mTOR downstream of PI3K-Akt. Such inhibition leads to decreased T and B cell activation and function as well as decreases cellular proliferation. Rapamycin has especially been used to treat advanced cancer [223,224] and transplant/graft-versus-host disease patients to mitigate rejection [225-227]. At the immune cell level, rapamycin can affect T cell differentiation and memory. Rapamycin treatment enhanced the quality and quantity of CD8 T cell memory responses by switching metabolism from glycolysis to oxidative phosphorylation [86]. Rapamycin can also mimic dietary restriction, increasing life span [228] as shown additionally in memory T cells [58,86]. Thus, caloric reduction and fat metabolism discussed above may be used as alternatives to rapamycin and may also improve T cell memory [60,229]. In CD4 T cells, rapamycin can instigate CD4 Treg development [230-232] as well as modulate chemokine receptors for mobilizing effector cells out of the periphery and back to the lymphoid organs [233]. The immunosuppressive effects of rapamycin, however, have been associated with higher infection rates [234,235] as well as hyperglycemia, pertaining to its inhibition of the PI3K-Akt-mTOR signaling pathway required for Glut1 expression and translocation to the plasma membrane [236,237], making it more suitable as a metabolic treatment for chronic inflammation and organ transplantation.

3-bromopyruvate is another anti-glycolytic drug preferentially used for the treatment of cancer. During glycolysis, a series of enzymes are necessary for the breakdown of glucose into pyruvate. 3-bromopyruvate is able to inhibit the activity of the first enzyme in the glycolytic pathway, hexokinase [238]. Hexokinase is often overexpressed in tumors and contributes to the high glycolytic activity seen in cancerous cells [24,25]. Treatment of tumor cells with 3-bromopyruvate drains intracellular ATP levels, resulting in cell death [239]. Although systemic treatment may be detrimental to actively proliferating and cytokine-secreting T cells, intra-tumoral treatment may allow for the induction of anti-tumor CD8 memory T cells, since a reversal of metabolism from glycolysis to oxidative phosphorylation is necessary for the maintenance of adaptive memory [59,89]. Additionally, 3-bromopyruvate is quite specific for cancer cells, with little to no toxicity of healthy tissue [240,241]. Therefore, eradication of the cancerous cell growth with 3-bromopyruvate may diminish the suppressive microenvironment surrounding the tumor, allowing for greater infiltration of cytotoxic killer cells and subsequent tumor immunity – an area of research that still requires investigation.

Anti-mitochondrial drugs

Blocking different mitochondrial complexes along the electron transport chain can manifest in phagocytic defects [242] and NF-κB inactivation [243]. These types of treatments are standard in lab settings; however, drugs that can be used in the clinic need to be better characterized. Metformin, for example, is an anti-oxidative phosphorylation drug and is predominantly used as an anti-diabetic drug. It enhances glucose disposal in muscle and reduces hepatic gluconeogenesis. Additionally, metformin was shown to enhance the levels of UCP2 [244]. Unlike rapamycin, which directly blocks mTOR, metformin can activate AMP-activated protein kinase (AMPK), which in turn blocks mTOR function, inhibiting cell proliferation [245,246]. AMPK activation induces glucose uptake from the blood, increasing glycolysis and lessening the dependence of cells on oxygen [247]. Metformin can specifically impede complex I of the electron transport chain and inhibit oxygen consumption in cancer cells [248,249]. Particularly, metformin usage for cancer therapy has been widely studied in those with type 2 diabetes [250,251], a known risk factor for tumor formation [252]. Like the anti-glycolytics mentioned previously, reductions in cancer cell growth may permit the cytotoxic killing of tumor cells by T cells, albeit when treatment is administered directly to the tumor. The impact of metformin has also been characterized in a mouse model of cancer vaccination. Upon metformin treatment of ova-specific T cells, a significant enhancement of memory CD8 T cells was detected along with better tumor regression. These results depict the ability of metformin to modulate the immune system outside of its anti-proliferative effects. Additionally, in the context of proinflammatory cytokines, metformin is able to block the activation of NF-κB, resulting in diminished cytokine-induced endothelial cell adhesion molecule expression [253,254]. Such an effect may lead to decreases in chronic macrophage infiltration into the adipocytes of type 2 diabetics as well as lower the damaging consequences of autoimmunity. Of course, better understanding of metformin’s selective effects on the immune system is necessary for these types of indications.

As another approach to metabolic control for immune regulation, antioxidants have been utilized in a plethora of disease models, including autoimmunity, infections, neurodegeneration, and cancer [255-259]. Although antioxidants do not specifically block oxidative phosphorylation, they are important for decreasing ROS produced from the electron transport chain during mitochondrial respiration, thus reducing damaging side effects. During aerobic respiration, the glutathione transported from the cytosol into the mitochondrial membrane is the only antioxidant available for metabolizing H₂O₂ [260]. Therefore, the augmentation of antioxidants present in immune cells may alleviate certain cellular dysfunctions. Specifically, antioxidants are able to show improvement of immune-mediated disorders, such as lymphocyte and macrophage function in the face of aging, septic shock, asthma, and type 1 diabetes [259,261-264]. The mechanisms by which antioxidants improve the immune system vary greatly. An antioxidant present in green tea called epigallocatechin-3-gallate can reduce T cell signaling via downregulation of cytokine receptors [265] as well as ameliorate EAE through an enhancement of regulatory T cells [266]. Other studies demonstrated the antioxidant ability of resveratrol to decrease collagen-induced arthritis by suppressing T_H17 responses [267] and to enhance B cell lymphoma recognition by CD4+ T cells through upregulation of HLA class II molecules [268]. Consistently, other antioxidants have shown promise in restricting NF-κB activation [269-271], leading to many anti-inflammatory effects. On the metabolic side, a recent study utilizing α-tocopherol, the antioxidant component of Vitamin E, demonstrated a reduction in glycolysis in lymphoma cells through blockade of lactate dehydrogenase activity [272].

Our group focuses on the usage of manganese porphyrins as catalytic antioxidants for the scavenging of ROS (superoxide, peryoxynitrite) as well as the mimicking of superoxide dismutase [259,273]. Unlike other antioxidants, the metalloporphyrins are catalytic and can repetitively eliminate ROS, resulting in many immunological effects. These antioxidants have shown promise in reducing type 1 diabetes incidence through autoreactive T_H1 cell modulation [259,274,275] and in protecting islets during isolation for transplantation [276,277]. Furthermore, NF-κB activation [271] and CD8 T cell effector function [278] are reduced upon metalloporphyrin treatment. Inhibition of NF-κB after metalloporphyrin administration has also shown promise in decreasing acute central nervous system injury, effectively enhancing neurologic function following ischemic stroke [279]. Additionally, metalloporphyrins can protect lungs from radiation-induced injury via HIF-1 inactivation [280-282] and kidneys from ischemia/reperfusion injury through induction of ATP synthase [283]. Interestingly, metalloporphyrins display oxidoreductase abilities, where they can act as a scavenger in the cytoplasm, as in the context of reduced TACE oxidation [274], yet work as an oxidizer in the nucleus, inhibiting the reduction of the p50 subunit of NF-κB and effectively blocking DNA binding [271]. In the context of cancer, metalloporphyrins are also able to block HIF-1 activation, decrease hypoxia, reduce tumor-protective cytokine release and ultimately suppress tumor growth [284]. HIF-1, as mentioned above, is critical for facilitating glycolysis in times of low oxygen; moreover, tumor cells rely heavily on glycolysis to survive, making them an obvious target of metalloporphyrin-induced regulation. The wide scope of metalloporphyrin effectiveness allows for their usage in a range of immunologic diseases, all centered around restoring redox balance; yet the effect of metalloporphyrins on fundamental immune cell metabolism has yet to be described (Figure 2). In conjunction with the cancer studies, we have observed promotion of hyporesponsive T cells after antioxidant treatment [274,285]. Preliminary studies suggest decreases in aerobic glycolysis following metalloporphyrin administration. This metabolic reduction may then decrease T cell differentiation and return cells to stasis or quiescence, all while retaining viability, as metalloporphyrins are not toxic [273,276,278]. If T cells are in fact displaying lowered aerobic glycolysis, the potential for treating chronic inflammatory conditions, such as autoimmunity, is widespread. With frontline therapeutics, such as anti-metabolites, rapamycin, and antibodies against costimulatory molecules, either failing in the clinic or leading to unwanted side effects, agents that modulate immune function, have mild side effects, and show no toxicity are highly sought after. Metalloporphyrins meet all of these demands and therefore, may be beneficial for reducing inflammatory disorders/potentiating cancer regression and should elicit greater attention in the search for alternative metabolic therapies.

Figure 2.

Metabolic modulation potential of manganese porphyrins. Manganese porphyrins, or catalytic antioxidants, have been shown to decrease NF-κB and HIF-1 activation, reduce T cell-mediated type 1 diabetes progression/onset, and maintain islet cell function during isolation and transplantation. Proposed metabolic effects of metalloporphyrins include decreasing: cell growth, hypoxia-induced reperfusion injury, the Warburg effect, and bioenergetics of diabetogenic immune cells.

Summary

The immune system plays a vital role in maintaining a fine balance in the battle against infections and cancer, but requires rigorous control in order to walk the fine line of regulator and menace leading to self-reactivity and autoimmunity. Both oxidative phosphorylation and glycolysis are critical for fulfilling the metabolic needs of immune cells. In innate immune cells, glycolysis is the predominant form of metabolism, whereas adaptive immune cells fluctuate between oxidative phosphorylation and glycolysis, depending on their activation status. Nonetheless, the heavy reliance of immune cells on glucose utilization makes them good targets for immunometabolic therapies. A number of endogenous molecules can be pursued, including HIF-1 and UCP2. Alternatively, caloric restriction, anti-glycolytics, and antioxidants all exhibit potential in resetting homeostasis in chronic inflammation while possibly enhancing immunity in cancer models. Overall, metabolic regulation should be an active line of research for the control of immune-mediated disorders.